On Dec 19, 2006, at 6:59 PM, Yoshiki Ohshima wrote:
Markus,
While ago, there was some discussion about adding a few tiles for controling the tone of sound. Apparently^^; you had some other high priority business back then. How things are going now? WorldStethoscope code is in the OLPC image now, and we are in the mood of enhancing the image toward the middle of January.
-- Yoshiki
Hi Yoshiki,
thanks for asking! The high priority business is done :-)) now I am back to business as usual...AND: There are holidays coming up now :-)
To summarize the "state of the art" on my side:
- I have a very basic csound etoy interface running here on my 1.67 non intel powerbook via osc which works without noticable latencies. But: I am "just" triggering samples right now, -- though with a very high precision when the samples should be triggered.
- So the delta of my spike solution with the stethoscope solution (if stethoscope didn't change since the last time I looked) is that I propose a tile where one could set _after which delay_ (say in terms of seconds) a sound with some frequency should be played. I think this funtionality is really useful as one could write little sequencers oneself. I made some funny toy projects where the x-axis of some morphs was bound to when the sound should be triggered and the y-axis to the freqeuncy of the sound.
- I have not fiddled yet with any sound-in capabilities of csound.
- I have some basic instruments/samples from TamTam working, but the frequencies and other stuff are not computed correctly yet, they need to be "normalized". I was playing with csd files (csound programs) directly and started there an osc loop waiting for some incoming osc events and triggering some notes. I am far away of being a "csd" hacker... So I just didn't do the maths involved here, and just made it -- well -- sound and not sound well... - So my big hope was to share some osc based python server, which is talking to csound, with the folks from TamTam -- I am just waiting for an answer about this subject from them. Simon?
I might have convinced them to use osc to do their client server functionality instead of sending some python commands over the line, as they did before and agreed with me to be insecure (at least if python is not sandboxed anyhow). As far as I have heared osc is used for the musical memory game already due to my suggestions for using this specific python osc library for TamTam...:-)
Bottom Line: To share my code with you now, you would need a csound version for the OLPC with osc build in.
The nicer solution would be to talk to a (probably python based) csound server via osc, as we could share a lot with the folks from TamTam who know better than most of us how to get some more funny beeps out of csound...
To get something going, could you build a csound version with osc? In any case I plan to package what I have done and publish it on squeaksource sometime during the next two weeks.
All the best,
Markus